Last Updated: 11/27/2011 by Corbitt Kerr 7/19/2011 by Corbitt Kerr This readme file describes the location of observed data for the Tropical section of the testbed. It also describes what the contents of each folder are. --------------------------------------------------------------------------------- For 2008-Ike and 2005-Rita Folders: Observation data has been collected in the form of 6 data types: hwm - high water marks watlev - waterlevel tp - peak period (SWAN_TPS) tm - mean period (SWAN_TMM10) dir - mean wave direction (SWAN_DIR) hs - significant wave height (SWAN_HS) Data was obtained from multiple sources. The orignal data varied in format and reference. In order to compare model data to observation data, all data was converted to SI, NAVD88, and UTC. In the case, where there was not enough information with regard to datums or time zones, certain assumptions were made to transform the data as appropriate. The process for conversion involved the following process: 1) Organize the data into folders identified by its source in the "Raw" folder directory. 2) Create a Stations file that listed the station id, filename, latitude, longitude, datum, datum adjustment (-9999 if NA), and datum adjustment source. 3) A fortran code was written to convert the raw data into IMEDS ascii format making adjustments (ft to meters, CDT to UTC, datum adjustments) as necessary. Note: There can be multiple formuations for wave data. Direction & Period were adjusted to conform to the same formulation used by SWAN(TMM10, TPS, DIR). The format of the IMEDS files was: TYPE_SOURCE.IMEDS (i.e. watlev_CRMS.IMEDS). 4) A fortran code was written to create a stations file from an IMEDS file. (i.e. watlev_CRMS.Stations) 5) The station file was used to obtain model data, which was created in IMEDS format. The output format was: TYPE_SOURCE.RUNID.MODEL.IMEDS (i.e. watlev_CRMS.Ike2D-wv.61.IMEDS , where 61 stands for ADCIRC). 6) A matlab code was written to compare multiple IMEDS files and output the figures to an internet based google maps. This required the conversion of the ASCII IMEDS to *.mat IMEDS. 7) Using the figures and the map, the locations of the stations were adjusted by creating a station overide file (*.L*). For example, the latitude and longitude provided placed the station location on the bank as opposed to the river. This file was used to overide the *.Stations file, with an *.L.Stations file. 8) Using the better Stations files, steps 5 and 6 were repeated. 9) Using the better figures and map, the observation data was cleaned, by flagging any data that was considered poor (i.e. flatline, plateau, dips, etc.). This was performed by creating a *.C.txt file in IMEDS format, but only lists those points that need to be flagged. The file also described the reasons for the flagging. This file is used to convert the watlev_CRMS.IMEDS file into a watlev_CRMS.C.IMEDS file. 10) While some data may be cleaned, some station data is entirely missing, poor, or unreliable. These stations are removed from the data set by creating a *.F.txt file that flags stations that need to be removed. These stations are removed from the set, leaving a filtered set called *.F.IMEDS. 11) Some models have the ability to interpret model data between nodes, however some do not. In order to ensure that all models are comparing data at the same station, a *.P.Stations file was created for each TYPE_SOURCE.IMEDS. This file lists a particular node near the latitude and longitude used for the station. It may feature the nearest or deepest node depending on the placement of nodes in the grid. This process was performed visually by overlaying station over the grid file in SMS and making selections. In Summary, there are summary files for each of the data types and two directories (Raw & Processed) in each of the storms folders. All Original Data was stored in the Raw folder. All Processed files (See Below) were stored in the Processed Directory. *.IMEDS - Observation data in IMEDS format *P.Stations - Preferred Node *.L.Stations - Adjusted Coordinates of Stations *.Stations - List of Stations *.F.* - Filtered Data (Selected Stations Removed) *.P.* - Preferred Nodes *.C.* - Cleaned Data --------------------------------------------------------------------------------- Listed in the Tides folder: This folder contains the data necessary for the tidal harmonic constituent comparison. NOAA_Harmonic_Analysis_Stns.txt has the station locations for NOAA stations to be compared to for tides. The archive NOAA_Harmonics_feet.tar has the harmonic data for the stations in feet and the archive NOAA_Harmonics_metric.tar has the harmonic data for the stations in meters. Note that both sets of data are set to GMT for time. !! There are now 60 stations of data instead of the 56 that were up at 11/01/2010.